The objective of this project is to investigate how protein-protein interactions between hepatitis C virus (HCV) and host-cell machinery contribute to viral infection and oncogenesis. This information can be leveraged to design broadly acting, resistance-proof inhibitors against HCV, while also informing efforts towards personalized therapy. HCV is a leading cause of liver cancer, however no vaccine exists against the virus. In addition, the utility of existing therapies is hindered by unpleasant side- effects, patient geneti heterogeneity (Slev, 2012), and rampant drug resistance (Welsch & Zeuzem, 2012). Many novel therapies against HCV are currently in clinical trials, but these, too, are likely to suffer from many of the same problems (Romano et al., 2012). Such challenges underline the need to personalize existing treatments, and moreover, to develop therapeutics that are robust to HCV evolution. This project works toward addressing both of these issues. In addition, it will explore whether protein- protein interactions with HCV incline the cell's evolutionary landscape towards cancer by masking the deleterious effects of pre-cancerous mutations. These objectives will be achieved by integrating evolutionary signatures (e.g. Patel et al., 2012); HCV+ hepatocellular carcinoma (HCC) sequencing data (Fujimoto et al., 2012); and 188 high-quality human-HCV protein-protein interactions from HEK 293T and Huh7 hepatoma cell lines, identified by collaborators in the Ott and Krogan labs. As a first aim, a classic test of selection (the McDonald Kreitman test) will be adapted to assess, for each gene and across a global range of populations, average levels of human-lineage adaptation. This will be supplemented by sitewise tests for selection across primates, and these signals will be compared between HCV-interacting proteins and the rest of the genome. Additionally, protein-protein interaction sets from other pathogens will be leveraged to elucidate HCV-specific effects. We expect this to provide useful information about how HCV co-ops host cell machinery. The positively selected sites that are identified in the previous step will then be used to define putative binding interfaces using the inferred or observed structure for each protein. Next, it will be tested whether HCV+ HCC-associated somatic mutations are enriched in these interfaces relative to the rest of the proteome. As a second aim, the important human-HCV protein-protein interactions identified in Aim 1 will be validated experimentally. For those highly selected proteins shown to interact with HCV, positively selected and control sites will be mutated. The effect of these changes on affinity for viral binding partners will be assessed by Surface Plasmon Resonance (SPR). Finally, we will conduct a similar analysis for mean viral titers by knocking down each endogenous gene of interest by shRNA, over-expressing the mutant form, and monitoring infection by a fluorescent strain of HCV.